1. Field of the Invention
The present invention relates to an image processing apparatus for performing an edge correction process which is applicable to a digital color copying machine, a color facsimile apparatus, an image filing system, and to a digital copying machine comprising the image processing apparatus therefor. More particularly, the present invention relates to an image processing apparatus for performing an edge correction process for image signals of respective pixels sequentially inputted, such as an edge emphasis process, and a digital copying machine for reproducing an image of a color document on a recording medium after reading the image of the color document and converting the read image into image signals, the digital copying machine comprising the image processing apparatus therefor.
2. Description of the Related Art
Recently, in an image processing apparatus such as a digital copying machine, it has been required to have various kinds of excellent functions. As these excellent functions for the image processing apparatus, for example, there have been provided a process for properly correcting an edge amount of an image signal converted by reading an image using an image reader, and also an edge emphasis process for increasing an edge emphasis amount of each edge included in an image such as a document including a map.
Documents to be copied using a digital copying machine often include a combination of black characters and a half-tone photograph. When copying a document of this combination, it is required to reproduce black characters having sharper edges without coloring them, and also it is required to reproduce a half-tone photograph with a better color reproduction characteristic or a better gradation characteristic which changes smoothly. In order to perform these processes, it is necessary to provide an area judgment process for certainly judging areas including black characters.
One example of a conventional edge emphasis process of a digital copying machine will be described below with reference to FIGS. 1 and 2.
FIG. 1 is a schematic block diagram showing a signal processing section of a conventional digital copying machine, and FIG. 2 shows a method for performing an edge emphasis process which is used in a conventional edge emphasis circuit 8 shown in FIG. 1.
Referring to FIG. 1, a document 1 is illuminated by an illumination lamp (not shown), and light including an image of the document 1 (referred to as a document image hereinafter) reflected from the document 1 is incident onto a CCD image sensor 3 through a focus lens 2. The CCD image sensor 3 converts the document image into an image signal including density information of the document image, and outputs the converted image signal through an amplifier 4 to an analogue to digital converter (referred to as an A/D converter hereinafter) 5, which converts the amplified analogue image signal into a digital image signal and outputs it to an edge emphasis judgment circuit 7 through a shading correction circuit 6.
The edge emphasis judgment circuit 7 judges whether or not an edge emphasis process is to be performed for the inputted digital image signal. If it is judged that the edge emphasis process is to be performed, the edge emphasis judgment circuit 7 outputs the inputted digital image signal to the edge emphasis circuit 8. On the other hand, if it is judged that the edge emphasis process is not to be performed, the edge emphasis judgment circuit 7 outputs the inputted digital image signal to an image processing circuit 9. In the judgment of the edge emphasis judgment circuit 7, when an absolute value of a difference between a density of a specified pixel and a density of each of plural peripheral pixels located within an area at a predetermined distance from the specified pixel is larger than a predetermined threshold value, it is judged that the edge emphasis process is to be performed. On the other hand, when the above absolute value is equal to or smaller than the above predetermined threshold value, it is judged that the edge emphasis process is not to be performed.
Further, the image processing circuit 9 performs an image process including a .gamma. correction process for the digital signal inputted from the edge emphasis judgment circuit 7 or the edge emphasis circuit 8, and outputs the processed image signal to a digital to analogue converter (referred to as a D/A converter hereinafter) 10 of a laser printer 20.
The laser printer 20 comprises the D/A converter 10, a laser modulator 11, a laser driver 12 and a laser diode 13 which are well known to those skilled in the art. The laser printer 20 reproduces the document image of the document 1 on a piece of copying paper in response to the digital image signal inputted from the image process circuit 9.
FIG. 2 is a schematic diagram showing a method for performing an edge emphasis process which is used in the conventional edge emphasis circuit 8 shown in FIG. 1. It is to be noted that the conventional edge emphasis process is disclosed in the Japanese patent laid-open publication (JP-A) No. 62-183672/1987.
Referring to FIG. 2, a specified pixel P0 is positioned in the center of a pixel window W3 composed of 3.times.3 pixels in a matrix form (referred to as a 3.times.3 pixel window W3 hereinafter). Four peripheral pixels P1 to P4 are positioned at the four corners of the pixel window W3. In the edge emphasis process, as shown in FIG. 2, the sum of the densities of the four peripheral pixels P1 to P4 is subtracted from the density of the specified pixel P0 multiplied by five so as to obtain an image signal for which the edge emphasis process has been performed. Namely, a secondary differential value of the density of the specified pixel P0 obtained by differentiating the density of the specified pixel P0 twice with respect to positions in a main scan direction and a subscan direction is subtracted from the density of the specified pixel P0, resulting in an image signal for which the edge emphasis process is performed.
In the edge emphasis process described above, there can be obtained such an effect that a noise included in the image signal is suppressed by the edge emphasis process. However, since there is a discontinuity between respective edge emphasis amounts of a pixel for which the edge emphasis process has been performed and a pixel for which it has not been performed, the reproduced image becomes an unnatural image.
Further, since the emphasis amount of the edge emphasis process is obtained by adding the above-mentioned secondary differential value calculated using fixed coefficients to the density of the specified pixel P0, there is such an disadvantage that the edge emphasis amount can not be freely changed. Furthermore, when the above-mentioned conventional edge emphasis process is performed for a color signal composed of color-dissolved signals of respective colors such as R, G and B signals, there is such a problem that the color of the image signal becomes an achromatic color if the level of the image signal is depart from a dynamic range of the image signal which is predetermined in the image processing apparatus.
One example of a conventional area judgment circuit used in a digital copying machine will be described below with reference to FIG. 3 which is a schematic block diagram of the conventional area judgment circuit. It is to be noted that the conventional area judgment circuit is disclosed in the Japanese patent laid-open publication (JP-A) No. 64-41377/1989.
Referring to FIG. 3, color signals R, G and B obtained by reading a document scanning it are inputted in parallel to a filtering process circuit 601 for a half-tone image, a filtering process circuit 602 for filtering a binary image and an area judgment circuit 610. The filtering circuit 601 is constituted by a two dimensional filter for performing a band emphasis process on the basis of such an assumption that the specified pixel is located in an area of a half-tone image. The frequency characteristic of the two dimensional filter is preset so that components of mesh points are rejected and the sharpness of the reproduced image is heightened. The filtering circuit 602 performs an edge emphasis process for edge components such as characters on the basis of such an assumption that the specified pixel is located in an area of a binary image. The binarizing circuit 603 binarizes the color signals R, G and B using respective predetermined threshold values, respectively, only when hue signals r1, g1 and b1 become a high level. On the other hand, when the hue signals r1, g1 and b1 become a low level, the binarizing circuit 602 outputs the color signals R, G and B as they are.
One set of color signals is selected by a selector 604 among a set of color signals filtered by the filtering circuit 601 and a set of color signals outputted from the binarizing circuit 603 according to an area judgment signal outputted from the area judgment circuit 610 which is described in detail later, and the selected set of color signals are outputted through a color correction and under color removal (referred to as a UCR hereinafter) circuit 611 to an image processing circuit of the next stage.
The area judgment circuit 610 comprises a hue judgment circuit 605, a ROM 606 for storing predetermined threshold values used for the area judgment, a signal combining circuit 607, an edge signal generator 608, and a comparator 609. The signal combining circuit 607 converts the inputted color signals R, G and B into luminance signals, combines the converted luminance signals and outputs the combined luminance signal d to the edge signal generator 608.
The edge signal generator 608 calculates the difference between the maximum value and the minimum value of respective densities of the pixels located within the 3.times.3 pixel window W3 having a specified pixel P0 positioned in the center thereof in response to the combined luminance signal d, and outputs an edge signal e having the level of the calculated difference to the comparator 609. The comparator 609 compares the edge signal e with the predetermined threshold value which is read out from the ROM 606. If the edge signal e is equal to or larger than the threshold value, it is judged that the pixel of the color signals R, G and B corresponding to the edge signal e is positioned in a binary image area including characters, and then, the comparator 609 outputs a high level area judgment signal to the selector 604. Otherwise, it is judged that the pixel of the color signals R, G and B corresponding to the edge signal e is positioned in a half-tone image area, and then, the comparator 609 outputs a low level area judgment signal to the selector 604.
The hue judgment circuit 605 judges whether or not the hue of the specified pixel P0 corresponds to either one of seven hues: yellow, magenta, cyan, black, red, green and blue, and outputs hue signals r1, g1 and b1 of three bits corresponding to the judged hue to the binarizing circuit 603 and the ROM 606. The hue signals r1, g1 and b1 are inputted to the address terminals of the ROM 606, and the ROM 606 outputs a digital signal of eight bits of the threshold value for the area judgment performed in response to the hue of the pixel, to the comparator 609. The comparator 609 judges whether the pixel is positioned in either a half-tone image area or a binary image area by comparing the edge signal e with the predetermined threshold value which is read out from the ROM 606, and outputs the judgment signal to the selector 604, as described above.
In order to select one set of color signals for which a process suitable for the kind of the image area where the pixel is positioned has been performed, one set of color signals is selected by the selector 604 among a set of color signals filtered by the filtering circuit 601 and a set of color signals outputted from the binarizing circuit 603 according to the area judgment signal outputted from the comparator 609, and the selected one set of color signals is outputted as color signals r, g and b to the color correction and UCR circuit 611. The color correction and UCR circuit 611 performs not only a color correction process for substantially rejecting the turbidity of the image of the pixel but also a UCR process, generates color signals C, M, Y and K of cyan, magenta, yellow and black, and outputs them to a color printer (not shown). The color printer reproduces a document image on a recording medium such as a piece of copying paper with toners of colors corresponding to the color signals C, M, Y and K in response to the color signals C, M, Y and K.
In the above-mentioned area judgment circuit shown in FIG. 3, when the area judgment circuit performs the area judgment process for a pixel positioned in a binary area having a slightly small edge component, it may be judged in error that a half-tone area having a slightly large edge component is a binary area. At that time, a half-tone image can not be reproduced with a gradation characteristic which changes smoothly. In order to prevent the above-mentioned erroneous area judgment, when the threshold values stored in the ROM 606 are changed, the area judgment circuit judges that only a binary image area having a sufficiently large edge component is a binary image area, resulting in deterioration of the reproductivity of the binary image area.
Further, taking a general characteristic of the color printer into consideration, when a character portion, particularly a black character portion which is important when reproducing a document image including a combination of characters and a half-tone photograph, is reproduced with black toner so as to form black characters, there can be obtained a better reproduction characteristic because of the following reasons. The color printer reproduces a full color image by forming toner images of cyan, yellow, magenta and black on a recording medium such as a piece of copying paper with superimposing them. However, when a black image is reproduced with toners of cyan, yellow and magenta by completely superimposing them, it is difficult to obtain a satisfied degree of achromatic color and a satisfied density. Further, if a position of a toner image of a color is shifted from a position of a toner image of another color upon superimposing plural toner images of above-mentioned colors, edges of each black character are colored with a color.